The function of natural killer (NK) and myeloid cells depends on the balance of a number of activating and inhibitory receptors. In the past few years, much effort has been focused on characterizing the structure and the function of the inhibitory NK cell receptors specific for MHC class I molecules. These receptors inhibit NK cell-mediated lysis of target cells that express class I MHC molecules, while allowing lysis of class I negative cells. NK cell inhibitory receptors, CD94 and NKG2, with its cytoplasmic immuno-tyrosine inhibitory motif (ITIM), defines a family of MHC class I molecule mediated negative/positive signaling receptors on the surface of NK cells. Both chains of the NK cell inhibitory receptors are members of the type II transmembrane C-type lectin family of receptors (CTLR). The ligand of this receptor family has been identified to be HLA-E, which presents the signal peptides of other class I MHC molecules. We have expressed and reconstituted several forms of CD94 and NKG2A extracellular ligand binding domains and determined the crystal structure of a CD94 homodimer. Another family of inhibitory receptors, termed Killer Immunoglobulin-like Receptors (KIR), recognize the classical class I MHC molecules. We have expressed a truncated form of a KIR receptor (2DL2)that recognizes HLA-Cw3 and determined the crystal structure of KIR2DL2 and HLA-Cw3 complex. KIR binds in a nearly orthogonal orientation across the alpha1 and alpha2 helices of HLA-Cw3, directly contacting positions 7 and 8 of the peptide. No significant conformational changes in KIR are observed upon complex formation. The receptor footprint on HLA overlaps with, but is distinct from that of the T-cell receptor. Charge complementarity dominates the KIR/HLA interface and mutations disrupting interface salt bridges seriously diminished binding. While most KIR contacts are to conserved HLA-C residues, a hydrogen bond between Lys 44 of KIR2DL2 and Asn 80 of Cw3 confers allotype specificity. KIR contact restricts P8 of the peptide to residues smaller than Val. A second KIR/HLA interface produced an ordered receptor- ligand aggregation in the crystal which may resemble receptor clustering during immune synapse formation. NKG2D, a member of CTLR superfamily and distantly related to NKG2A, B, C, and E, is found on NK cells to trigger cytotoxicity against certain tumor cells and on CD8+ T cells to provide a co-stimulatory signal against virally infected cells. More recently, a group of human cytomegalovirus glycoproteins UL16 binding proteins, named as ULBPs, have been identified as ligands for human NKG2D. We have expressed both NKG2D and its ligands ULBP2, and ULBP3. We also completed the structure of a NKG2D/ULBP3 complex at a 2.6 angstrom resolution. In the NKG2D/ULBP3 complex, the structure of ULBP3 resembles that of the a1 and a2 domains of classical MHC molecules without a bound peptide. The lack of a3 and b2m domains is compensated by replacing two hydrophobic patches at the underside of class I MHC beta-sheet floor with a group of hydrophilic and charged residues in ULBP3. NKG2D binds diagonally across the ULBP3 helices, creating a complementary interface, an asymmetrical subunit orientation and local conformational adjustments in the receptor. The interface is stabilized primarily by hydrogen bonds and hydrophobic interactions. Unlike the KIR receptors that recognize a conserved HLA region by a lock-and-key mechanism, NKG2D recognizes diverse ligands by an induced-fit mechanism. Natural cytotoxicity receptors (NCR) are the major receptors responsible for NK cell-mediated lysis of tumor and viral-infected cells. Recent cloning of three NCRs, NKp46, NKp44, and NKp30, prompted studies by a number of investigators to elucidate their function as well as to identify their ligands. Despite these efforts, the ligands recognized by NCR remain unknown. To gain insights into the molecular mechanism of NCR-mediated NK cell activation, we have determined the crystal structure of human NKp46. The structure revealed striking resemblence between NKp46 and KIR receptors and enabled us to propose a potential ligand binding site. In addition, we also solved the crystal structure of the putative ligand binding domain of TREM-1, a myleoid cell activating receptor.